0521571766 cambridge university press language and social relations dec 2006

LANGUAGE AND SOCIAL RELATIONS

Language connects people to each other in social relationships and allows
them to participate in a variety of activities in everyday life. This original
study explores the role of language in various domains of our social life,
including identity, gender, class, kinship, deference, status, hierarchy, and
others. Drawing on materials from over thirty languages and societies, this
book shows that language is not simply a tool of social conduct but the
effective means by which human beings formulate models of conduct.
Models of conduct serve as points of reference for social behavior, even
when actual conduct departs from them. A principled understanding of the
processes whereby such models are produced and transformed in large-scale
social history, and also invoked, negotiated, and departed from in smallscale social interactions provides a foundation for the cross-cultural study of
human conduct.
A S I F A G H A is Associate Professor in the Department of Anthropology at the
University of Pennsylvania, and editor of The Journal of Linguistic
Anthropology.


STUDIES IN THE SOCIAL AND CULTURAL FOUNDATIONS
OF LANGUAGE

The aim of this series is to develop theoretical perspectives on the essential social
and cultural character of language by methodological and empirical emphasis on
the occurrence of language in its communicative and interactional settings, on the
socioculturally grounded ‘‘meanings’’ and ‘‘functions’’ of linguistic forms, and on
the social scientific study of language use across cultures. It will thus explicate the
essentially ethnographic nature of linguistic data, whether spontaneously occurring
or experimentally induced, whether normative or variational, whether synchronic
or diachronic. Works appearing in the series will make substantive and theoretical

contributions to the debate over the sociocultural-function and structural-formal
nature of language, and will represent the concerns of scholars in the sociology and
anthropology of language, anthropological linguistics, sociolinguistics, and socioculturally informed psycholinguistics.
Editors

Editorial Advisers

Judith T. Irvine
Bambi Schieffelin

Marjorie Goodwin
Joel Kuipers
Don Kulick
John Lucy
Elinor Ochs
Michael Silverstein

A list of books in the series can be found after the index.


LANGUAGE AND SOCIAL
RELATIONS

ASIF AGHA
University of Pennsylvania


CAMBRIDGE UNIVERSITY PRESS

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo

Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521571760
© Asif Agha 2007
This publication is in copyright. Subject to statutory exception and to the provision of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.
First published in print format 2006
eBook (NetLibrary)
ISBN-13 978-0-511-25723-0
ISBN-10 0-511-25723-6
eBook (NetLibrary)
ISBN-13
ISBN-10

hardback
978-0-521-57176-0
hardback
0-521-57176-6

ISBN-13
ISBN-10

paperback
978-0-521-57685-7
paperback
0-521-57685-7

Cambridge University Press has no responsibility for the persistence or accuracy of urls
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.


CONTENTS

List of figures
List of tables
Acknowledgments
Typographical conventions
Introduction
1 Reflexivity
1.0 Introduction
1.1 Reflexive activity
1.2 Text-level indexicality and interactional tropes
1.3 Reflexive activity in interaction
1.4 Deixis and representation
1.5 Performativity
1.6 Reflexive processes across encounters
1.7 Large scale cultural formations

page viii
x
xiii
xv
1
14
14
16

24
27
37
55
64
77

2 From referring to registers
2.0 Introduction
2.1 Referring
2.2 Propositional stance and role alignment
2.3 Denotational categories
2.4 Norms of denotation and interaction
2.5 Dialect, sociolect and denotational footing
2.6 Retrospect and prospect

84
84
85
96
103
124
132
142

3 Register formations
3.0 Introduction
3.1 Three aspects of registers
3.2 Metapragmatic stereotypes
3.3 Stereotypes and socialization

3.4 Stereotypes and ideology
3.5 Entextualized tropes
3.6 Fragmentary circulation

145
145
147
150
155
157
159
165
v


vi

Contents
3.7
3.8
3.9
3.10
3.11

Reflexive social processes and register models
Sociological fractionation and footing
Semiotic range
The enregisterment of style
Conclusion

167
171
179
185
188

4 The social life of cultural value
4.0 Introduction
4.1 Received Pronunciation: basic issues
4.2 Metadiscourses of accent
4.3 The emergence of a standard
4.4 The transformation of habits of speech perception
4.5 The transformation of habits of utterance
4.6 Asymmetries of competence and perceptions of value
4.7 Changes in exemplary speaker
4.8 The sedimentation of habits and the inhabitance of agency

190
190
191
195
203
206
219
223
224
228

5 Regrouping identity
5.0 Introduction

5.1 From ‘identity’ to emblems
5.2 Emblematic figures of identity
5.3 Role designators and diacritics
5.4 Emblematic readings
5.5 Interaction rituals as emblems of group status
5.6 Emergent, stereotypic and naturalized groupings
5.7 Enregistered identities and stereotypic emblems

233
233
234
237
246
254
260
268
272

6 Registers of person deixis
6.0 Introduction
6.1 Metapragmatic stereotypes and standards
6.2 Reflexive processes within pronominal registers
6.3 Emblems of social difference
6.4 Troping on norms
6.5 Social boundaries

278
278
279
286

293
295
298

7 Honorific registers
7.0 Introduction
7.1 Variation and normalization
7.2 Lexeme and text
7.3 Pronominal repertoires
7.4 Phonolexical registers of speaker demeanor
7.5 Registers of referent-focal deference
7.6 Deference to referent: text-defaults

301
301
302
304
308
310
315
317


Contents
7.7 Textually composite effects
7.8 Social domain
7.9 Speech levels
8 Norm and trope in kinship behavior
8.0 Introduction
8.1 From kinship systems to kinship behavior

8.2 Lexicalism, codes, and the genealogical reduction
8.3 From kinship terms to text-patterns
8.4 Normalized tropes
8.5 Renormalization and standards
8.6 Society-internal variation
8.7 Sign and metasign in kinship behavior
8.8 From cultural kinship formations to any cultural
formation
Notes
References
Index

vii
322
332
334
340
340
341
346
350
356
368
372
375
382
386
408
419

FIGURES

1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
3.1
3.2
3.3
3.4
3.5
4.1
4.2
viii

Metasemiotic motivation of icons
Metasemiotically motivated co-occurrence effects:
text-level indexicality
Reflexive descriptions of speech co-text and context
Self-reported strategies for modeling next-turn behavior
(Swedish)
Communicative transmission through a speech chain
Communicative networks in mass communication
Dyadic conversation
A biographic history of encounters
Soliloquy and inner speech
Referential vs. attributive uses
Text configurations as referring signs
Multi-channel sign-configurations and participant
alignments
Structural sense classes of English noun
Deictic signs: denotational and interactional schemas
Denotational stereotypes as social regularities
Referential prototypes
Reanalysis of Thai syntactic patterns into ‘high’ and ‘low’
registers
Role configuration and denotationally-mediated
footing
Three levels of engagement with register phenomena
Gender tropes in Lakhota
Register-mediated alignments: reanalysis and
self-differentiation
Javanese Wayang Kulit: ritual comportment as implicit
typification
Calvin and Hobbes on registers, voicing, tropes and

recirculation
Bateman cartoon, 1920
Diagrammatic motivation of co-occurring variables

page 22
24
32
35
67
69
70
70
71
92
93
101
114
118
121
122
130
138
149
161
173
184
188
198
211

Figures
5.1
5.2
5.3

Role designators and diacritics
Textually cumulative models of personhood
Title page, Elements of Elocution by John Walker,
1806 edition

ix
249
252
274


TABLES

1.1
1.2

Propositional content and propositional act
Non-selective deixis: from definite past queries to
‘nomic’ truths
1.3 Categorial text-defaults for common English deictics
1.4 Categorial text-defaults and cross-linguistic comparison
1.5 Co-textual specification of denotational schema
1.6 Role fractions of speaker
1.7 Predicate modalization and the explicit performative

locution
1.8 Cross-cultural comparisons of performative locutions
2.1 Reflexive evaluations presupposed in choice of
location-referring term
2.2 Denotational text as a performed diagram of
interactional text
2.3 Distributional structure and sense categories
2.4 Structural sense classes of suffix
2.5 Thresholds of normativity
2.6 Tamil caste sociolects and denotational footings
2.7 Register contrasts in Norwegian
3.1 Typifications of language use
3.2 Phonolexical registers of speaker gender in Koasati
3.3 Lexical registers of speaker gender in Lakhota
3.4 Some features of ‘Sports announcer talk’
3.5 American military register
3.6 Some dimensions of register organization and change
3.7 Mirror-image alignments in Egyptian Arabic: a case
of inverse iconism
3.8 Reanalysis in stereotypic values of second person polite
pronouns (Javanese)
3.9 Javanese kasar ‘coarse’ vocabularies
3.10 Javanese kasar ‘coarse’ prosody
3.11 Multi-channel indexical icons in Javanese
4.1 Phonolexical changes in RP vowels: selected examples
x

page 41
43
47

48
49
50
57
62
95
99
112
115
126
137
139
151
160
160
164
166
169
174
176
181
182
183
194


Tables
4.2
4.3
4.4

4.5
5.1
5.2
5.3
5.4
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
6.10
7.1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
7.9
7.10
7.11
7.12
7.13
7.14
7.15

7.16

Popular media mis-spellings of U-RP words
Accent speech levels in Britain
Patterns of accent evaluation and role alignment
in West Wirral
Rough estimates of genre circulation
Greeting form as emblem of speaker’s status
Emblems of piety and honor
Processes that operate over role diacritics and emblems
Enregistered indices of refinement linked to ‘Received
Pronunciation’
Common genres of metapragmatic discourse about
pronouns
Person-referring pronominal forms in Thai
Pronominal registers in European languages
Text configurations marking politeness in French
Reanalysis of Italian second-person pronominal address
Levels of second person pronominal deference in Maithili
Second person pronominal levels in Urdu
Honorific nouns used as ‘pronouns’ in Urdu
Norms of appropriateness reported for Yiddish address
Social groups and dispensation rights in Swedish
Metapragmatic data used in the study of honorific
repertoires
Malayalam honorific ‘pronouns’
Phonolexical registers in Samoan
Enregistered phonolexical styles in Standard Teherani
Persian
Referent-focal deference: a cross-linguistic approximation

Cross-linguistic distribution of lexemic honorific forms
Honorific forms whose appropriate use depends on two
interactional variables
Conjoined focus in Japanese verbs of giving
Textually ‘superposed’ effects and interactional tropes
Javanese [Àhuman] common nouns
Addressee honorifics in Javanese: repertoire
contrasts
Korean speech levels: the early Standard system
Korean speech levels: contemporary Cihwali
Lexeme cohesion and speech levels (Tibetan): five ways
of saying ‘Mother went to the house’
Javanese speech levels: early Standard
Acceptability of ‘mixed’ speech levels by social domain
of evaluator: ‘conservative’ vs. ‘modern’ speakers

xi
197
200
201
217
263
266
272
273
281
285
287
287
289

290
290
292
296
299
306
309
311
313
318
321
330
331
332
333
334
335
336
337
338
339


xii
8.1
8.2
8.3
8.4
8.5
8.6

8.7
8.8

Tables
Common patterns of addressee-referring and third person
kinterm usage
Role inhabitance and referential gloss (English)
Role inhabitance and referential gloss (Vietnamese)
Address inversion in Japanese: referring to self as
addressee would or should
Recentered affinal address (Bengali): addressing others
as someone else would
Address recentering in Japanese
Patterns of kinterm usage
Address recentering in Hindi

352
354
355
359
361
362
365
374


ACKNOWLEDGMENTS

A first draft of this book was completed while I was a Fellow at the
Stanford Center for Advanced Study in the Behavioral Sciences during

2003–04. In addition to the opportunity to work full time on the book, the
Center and its extraordinary staff helped create conditions for writing and
reflection that were no less than ideal. Faculty grants from the University
of Pennsylvania and the University of California, Los Angeles made
possible sabbatical leaves at earlier stages of writing. I am immensely
grateful to these institutions for their support.
Conversations with other Fellows at the Center, especially Thomas
Welskopp, Walter Johnson, Jane Hill, and Webb Keane, remained a
daily source of intellectual stimulation while the book was being written.
Judith Irvine, Stanton Wortham, Douglas Glick, Robert Moore, Irene
Applebaum, Sabina Perrino, Michael Lempert, Luke Fleming and
Constantine Nakassis read the first draft of the manuscript and responded
with comments. At an earlier stage, James Kurichi provided insights into
Malayalam and Ed Keenan into Malagasy. Bob Agajeenian and Andrew
Schwalm worked as my research assistants. I am grateful to them all.
Thanks also to my two Cambridge editors, Judith Ayling, who first
invited me to write a book proposal and then accepted it, and Andrew
Winnard, who, with a kindness bordering on friendship, saw it through to
the form that now lies in your hands.
Some of the material in this book has been presented at various conferences over the years, and earlier versions of parts of the argument have
appeared in print. The basic view of registers underlying Chapter 3 was
first presented at a panel organized by Alessandro Duranti at the American
Anthropological Association’s annual conference in November 1997; a
portion of the chapter (perhaps two-thirds of the current version) was
published as Agha 2004 in an anthology that eventually emerged from
this panel. Different portions of Chapter 4 were presented at the Ethnohistory seminar at the University of Pennsylvania in November 1999, at a
panel organized by Greg Urban at the American Anthropological
Association in November 2000, and at the Anthropology Department
Colloquium of the University of Chicago in March 2002. I’d like to thank
xiii

xiv

Acknowledgments

Ben Lee, Richard Bauman and Sue Gal for stimulating comments during
these presentations and Michael Silverstein for comments on an earlier
version, published as Agha 2003. A portion of Chapter 7 was first presented
to the Japanese Association of Sociolinguistic Sciences in March 2002 and
published in their proceedings as Agha 2002. I thank Sachiko Ide and
Kuniyoshi Kataoka for the invitation to present my work at the conference and for their extraordinary hospitality during my visit. About half the
material in Chapter 8 was first presented at the Symposium About
Language in Society at the University of Texas, Austin, in April 2005,
and is scheduled to appear in their proceedings. I am immensely grateful to
the organizers of these symposia and to members of audiences for feedback
and discussion.
This book is dedicated to my son Omar Sheheryar Agha, now 11, who
helped me find examples of register phenomena in Calvin and Hobbes
comic strips, and, when they were found, helped me find lots of other
things too, so many that even Calvin and Hobbes had some trouble
guessing what they were.


TYPOGRAPHICAL CONVENTIONS

I use boldface
For technical terms when first introduced and occasionally thereafter to
remind the reader of their technical senses
‘Single quotes’

1. For glosses of expressions and utterance-acts
2. For quotations from authors (except when numbered and set on a
different line)
3. For everyday usages and terminologies on which I wish to comment
Italics
1. For forms of words and expressions in orthographic representation
2. For expository emphases
‘‘Double quotes’’
To clarify levels of embedding in reported speech
As for the linguistic data cited in this book, I have used IPA conventions
whenever possible, but have left intact the conventions used by the many
authors I cite when these depart from them.

xv



INTRODUCTION

Social relations vary across human societies in ways that are limitlessly
varied, endlessly susceptible to reanalysis, periodic stabilization and
change. Yet they are highly systematic in each locale for persons who
recognize themselves as so related. The goal of this book is to show that
such possibilities of variation and change, and their actual determinacy for
particular social actors, can only be explained given an adequate conception of the role of language in human affairs. Doing so requires that we
move beyond a variety of folk-views of language that exist among its users
in particular times and places; for instance, that language is primarily a
collection of words; that language is abstract, mental, devoid of materiality; that it stands apart from the ‘things’ that it inertly represents. We will
be building towards a rather different conception of language here, a view
that focuses on the materiality of language and its relationship to other

material things, on classifications of behavior that can be inhabited
through behavior, and on processes whereby classifications of behaviors,
and of those whose behaviors they are, can be maintained or modified
within the order of social interaction in which they are experienced.
It has often been supposed that the variability of social relations
observed across societies and history can be tamed by means of various
top-down approaches, as in the creation of taxonomies of ‘kinds of society’
viewed as explanations of what people do; or by enumeration of ever more
abstract cognitive universals believed to constitute structures of mind
independent of human action; or by resort to principles of functional
explanation through which actions tend to certain equilibria and yield
particular social formations as homeostatic results. There is no difficulty
even today in making up such stories about society. The difficulty is,
rather, that in order to appear plausible such accounts must ignore vast
realms of human experience attested in the ethnographic and historical
record, or harness such variation to evolutionist metaphors, or lay claim to
the greater rationality of their own moment in the history of the human
experiment even as this moment slips away.
This book builds in a different direction. I argue that the organization of
social life is shaped by reflexive models of social life, models that are made
1


2

Language and social relations

through human activities and inhabited through them, though not always
by the same persons. If the term ‘model’ seems a bit abstract there are many
other terms – idea, image, discourse, position, response, habit, ideology,

practice – that are variously appropriate in its place. All these terms convey
the notion of an enacted representation, a thing made somewhere through
some activity conveying something about another. One of the curious things
about language is that it allows us to formulate models of phenomena that
are highly abstract, even timeless; one of the curious things about our folkviews of language is their tendency to neglect what is obvious to our senses,
namely that any such representation, however general in import, must be
conveyed by a perceivable thing – i.e., be materially embodied – in order to
become known to someone, or communicable to another. These moments
of being made, grasped, and communicated are the central moments
through which reflexive models of language and culture have a social life
at all. And persons who live by these models (or change them) do so only
by participating in these moments.
These moments are of focal interest throughout this book. This focus
does not replace other concerns. It orients them. I discuss a large number
of traditional topics in this book, matters of longstanding interest to
students of language, culture and society. But I propose that careful
attention to such moments of making and unmaking allows us to solve
many of the most vexing problems we face in conceptualizing our subject
matter. Despite the fact that some reflexive models of human behavior
perdure or persist through time, some even for a long while, and despite the
fact that some among them persist through arrangements that formulate
them as timeless, exceptionless, essential, dominant, and so on, the central
and inescapable fact about human societies is the diversity of reflexive
models of behavior that co-exist within each society (and thus across
societies) at any given time. This diversity is partly a result of the fact
that persons have interested stakes in – they seek to own, disown, maintain
or re-evaluate – the models by which they live, though it has other sources
too. Such diversity is the taxonomist’s nightmare. But this is as it should
be, because, when it comes to culture, taxonomy is taxidermy.
Our goal here is to consider culture as a living process, as a thing whose

arrangements are continually renewed – though not always at the same
rate, or all at once – through the form-giving fire of human activities. The
notion of activity relevant here is semiotic activity – the use of enacted
representations in the sense discussed above – through which reflexive
models of behavior are made, inhabited, and re-made by the semiotic
labor of persons oriented to historical institutions. In many ways, this
book is an attempt to argue that human activities yield material precipitates and projections (things made through activity, ‘artifacts’ of various
kinds) that carry semiotic value or significance to those who perceive


Introduction

3

them. This point is fairly obvious for the case of durable artifacts. Yet
human beings make artifacts of different degrees of durability, whose
cultural meanings and consequences persist for different scales of time. If
human beings are artifact makers, the artifacts they most readily make are
enacted representations, including utterances and discourses. As individuals, we do this countless times a day and think nothing of it; but those
patterns of individual activity that we call institutions do it in a more
complex, sometimes puzzling way, and often with far greater consequence.
It is therefore all the more important to see that utterances and discourses
are themselves material objects made through human activity – made, in
a physical sense, out of vibrating columns of air, ink on paper, pixels in
electronic media – which exercise real effects upon our senses, minds, and
modes of social organization, and to learn to understand and analyze these
effects. It is true that utterances and discourses are artifacts of a more or
less evanescent kind (speech more than writing). But these are questions of
duration, not materiality, and certainly not of degree or kind of cultural
consequence. Things that last for seconds can have effects that last for

years. Even physical tokens of discourse that have a fleeting durational
existence (such as spoken utterances) can order and shape social relations
of a much more perduring kind, ones that persist far longer than the initial
speech token itself, whether through uptake in the subsequent activities of
others, by incorporation into widely routinized practices that rely on and
replay them, or by conversion into artifacts of a more durable kind. Every
argument in this book assumes the materiality of language and other signs.
But I reject the privileged status typically accorded in contemporary
discussions of materiality to the narrow special case of durable objects.
Such an emphasis, which fixates on the physical persistence of the durable
object, obscures the processes through which its sign-values emerge or
change. Last year’s hat doesn’t make the same fashion statement this
year. It’s the same hat. Or is it? Everyone agrees that fleeting signs (such
as spoken utterances and gestures) acquire contextual significance from
their more durable physical setting. It remains to be seen that the semiotic
values of durable objects (the kinds of things one can put on the mantelpiece, or trip over in the dark) are illuminated for their users by discourses
that appear evanescent even when their effects are not. In this book,
I attempt to make clearer attributes of language that shape the significance of perceivable objects across thresholds of durability in various
ways, whether by allowing fleeting signs to borrow significance from
ones that persist, or vice versa, or by making evanescent sign-values
more durable, or by causing enduring cultural phenomena to fade into
disrepute and disuse. It will soon become clear that many of these
attributes make language so exquisite an instrument for doing work – for
acting and interacting, for making and unmaking, for imbuing objects


4

Language and social relations

(including discourse itself) with value – that its products, or ‘works,’
are far more accessible to our everyday awareness than the instrument
itself.
Chapter 1 introduces basic concepts of reflexive activity, its varieties,
and a way of conceptualizing the scales of sociohistorical process in which
its effects (products, models, ‘works’) are experienced. Chapter 2 develops
themes pertaining to the issue of enacted representation, the character of
acts of referring (to ‘things’) as interpersonal achievements, the sociology
of denotation, and the normativity and authority of forms of representation. Chapter 3 develops an account of register formations, viewed now as
systems of socially significant signs (involving language and non-language)
that are formed, maintained, and reanalyzed through reflexive activities.
The account presented in these three chapters expands our conception of
what a register is (beyond the traditional view that registers are sets of
socially valued words and expressions) to a model where the kinds of signs
that comprise registers, the processes of valorization that establish their
sign-values, and the persons for whom they function as signs are all shown
to be features of a register not fixed once and for all but variables whose
values are defined and negotiated through reflexive processes within social
life. These aspects of the model allow us to conceptualize register formations as cultural models of action, as stereotypic ways of performing ‘social
acts’ of enormous range and variety, a variety exhibited not merely in their
intelligible social consequences but also in the range of phenomenal behaviors in which they are embodied.
Chapter 4 develops an account of enregisterment, the process whereby
one register formation comes to be distinguished from other modes of
activity, including other registers, and endowed with specific performable
values. Whereas all the other chapters in the book take a comparative look
at phenomena in different languages and societies, the comparative focus
of Chapter 4 is on different historical periods of a single language/society.
The next few chapters examine different types of enregistered signs.
Chapter 5 focuses on the social logics that underlie enregistered emblems
of ‘identity,’ and on matters of self- and other-positioning that emerge out

of these logics. Chapters 6 and 7 take up honorific register formations,
cases where enregistered signs are linked in ideologically explicit ways to
matters of respect, status, power and rank. Chapter 8 discusses processes
of enregisterment that bear on matters of kinship. The chapter illustrates
the enormous range of interpersonal relations that can be established
through kinship behaviors (the use of kinterms and associated non-linguistic
signs), both behaviors that conform to norms of kinship and those that
trope upon them. Behaviors of the latter kind establish forms of propinquity that are ‘kinship-like’ only in certain respects, but which, through
further processes of reflexive reanalysis, can be re-evaluated as new norms


Introduction

5

of kinship for certain social purposes, thereby resetting the standard to
which further analogues of kinship are referred.
This dialectic of norm and trope is central to social processes discussed
throughout this book. The sense in which social processes are limitlessly
varied, as I claimed in my opening sentence, is not that they vary randomly
or that ‘anything goes.’ This is far from the case. To see this we have to
recognize two distinct issues. First, although cultural models are often
normalized by social practices so as to constitute routine versions of
(even normative models for) the social behaviors of which they are models,
they can also be manipulated through tropes performed by persons
acquainted with such models to yield variant versions, and the range of
these tropic variations is potentially limitless. The second point is this. The
existence of cultural models and tropic variants also involves sociological
asymmetries. Not all norms that exist in a society are recognized or
accepted by all members of that society. Similarly, not all behaviors that

trope upon norms occur equally routinely or are intelligible equally widely;
not all intelligible tropes are ratified by those who can construe them; not
all the ones that are ratified come to be presupposed in wider social
practices, or get normalized in ways that get widely known. Each of
these asymmetries imposes some further structure on the first process
I described. I argue in this book that if we understand this dialectic of
norm and trope in semiotic terms, and if we know how to study these
asymmetries in sociological terms, the fact that cultural models vary in
(potentially) limitless ways is no cause for distress. Rather, a recognition of
this fact and the ability to explain its consequences helps us to understand
better the sense in which culture is an open project, the ways in which
forms of social organization are modifiable through human activities, and,
through a recognition of the various ‘positionalities’ generated by these
asymmetries, to recognize that the processes whereby cultural variation
comes about make untenable any form of radical relativism that presumes
the perfect intersubstitutability of social ‘positions.’
I use the expression ‘a language’ in this book to refer to the kinds of
phenomena to which we ordinarily refer by means of words like French,
Chinese, Arabic, or Tagalog. The term has no further technical specificity.
None is needed since more precise claims about reflexive processes are
formulated in the terminology of sign-functions introduced in Chapter 1.
When I use the generic term ‘language,’ my intent is to say: Pick any
language that you like. But I do not use this term for what is called
‘Language’ by some linguists (‘grammar’ will do here; more on this below);
if my arguments prove persuasive, the epistemological status of the capital-L
construct will need to be re-thought. I specifically refer to matters of
grammar and grammatical organization by using those terms. Other
more specific terms like ‘dialect’ and ‘sociolect’ are introduced in the text.

6

Language and social relations

A different set of considerations apply to the term language ‘use.’ The
term is an imperfect way of talking about events of semiosis in which
language occurs. As we examine the orderliness of such events we find
that there are several ways in which the unity of this construct, this thing
called language ‘use,’ breaks down. First, the term ‘use’ is itself ambiguous
between an act of performing an utterance and an act of construing it; here
‘use’ breaks down into ‘performance and construal’ or ‘act and response.’
Second, to say that language is being used is generally to point to the fact
that an array of signs is being performed and construed by interactants, of
which language is but a fragment; when language occurs in ‘use,’ it occurs
typically as a fragment of a multi-channel sign configuration, whose
performance and construal, enactment and response, constitutes the minimal, elementary social fact. Third, much of what is traditionally called
the data of ‘usage’ by linguists and others consists, in fact, of the data
of reflexive models of usage (e.g., norms and standards of usage) to which
the actual practice of using language does not always conform even in the
society where such data are gathered. These issues require that we distinguish different varieties of usage – an instance of usage, a habitual usage, a
normative usage, a tropic usage – in conceptualizing the kinds of work that
is accomplishable through language itself.
This book presents methods and frameworks for analyzing many
aspects of language. I offer extended discussion of examples from a variety
of linguistic and sociohistorical locales, relying on the work of many
others. Many of these data are summarized in tables, with source authors
and texts indicated at the bottom of the table. At various points in the
exposition I have found it convenient to highlight certain features of the
argument by setting them off from the text as summaries of the discussion.
These are cross-referenced in the text with a preceding S for summary by

chapter and summary number (as S 1.1, S 1.2, etc., in Chapter 1, and
so on). I have tended to highlight by way of summary those features of the
discussion in a particular chapter to which discussions in other chapters
make reference. The intention is to provide pointers and flags foregrounding a few selected themes so that the reader can re-visit issues which
animate discussions elsewhere in the book. In all cases the summaries
offer synopses of points discussed and exemplified at greater length in
the body of the text. But they differ among themselves in other respects. In
most cases the summaries occur immediately after the discussion summarized. In a few cases, they highlight themes preemptively, offering synopses
of materials that follow in the next two or three pages. In one or two
instances the summary highlights issues discussed in a previous chapter in
order to formulate a bridge or connection to the material now at hand.
Although these summaries always offer a synopsis of issues illustrated
by examples, they sometimes state synopses in formulations more general


Introduction

7

than local examples appear to warrant; this is invariably because the local
examples are instances of a more general phenomenon, of which additional examples from many languages and societies, cross-referenced to the
summary, occur later. So whereas all of these summaries have a common
expository function (that they are synopses of local parts of the text) they
are also variously, and additionally, flags, pointers, connectors, bridges to
other parts of the text, and sometimes generalizations which unite together
different portions of a more extended argument. The reader may be able to
use these summaries in various ways. But they are not intended as selfstanding claims isolable from the empirical cases which furnish their point,
nor as adipose verities of some armchair theory in which we may come to
find some everlasting rest (which is when they would become most adipose).
A great deal of ink has been spilled in the last forty years in pursuing the

assumption that the study of language is the study of ‘rules’ or ‘constraints’
on language. As with any fad, the time for this one has come and gone.
There is a simple trick that forms the basis for – and explains the popularity
of – the fad. The trick itself has two parts. Here’s how to do it. First,
redefine what the word language means, preferably fixating upon a fragment or feature of language – let’s say the concatenation system of language, its syntactic and phonotactic aspects – and call this fragment
‘language’ (or even ‘Language’). Second, redefine the study of this fragment as the study of some restricted type of data about it, let’s say the study
of decontextualized intuitions about it. If you’ve done this carefully
enough, you can now amaze and amuse your friends by pulling a vast
number of rules and constraints out of the hat of introspectable intuitions.
And, now, the statement ‘the study of language is the study of constraints’
appears to be true. But a more accurate way of stating this truth is ‘the
study of decontextualized intuitions can isolate plenty of features of a
concatenation system that appear as inviolable constraints to those intuitions.’ You can also do this for discourse. So, in your first step, you can
redefine ‘discourse’ as some genre of discourse, let’s say ‘conversation.’
And in your second step, you can define your privileged data type as
‘transcripts of conversation.’ You can now come up with all kinds of
formalizable constraints on discourse itself – the examples are right
there, after all, in those very transcripts! – and appear to prove that the
study of discourse is the study of constraints on conversation structure as
long as you don’t worry about the question: For whom?
Suppose now that someone else does this, and you are part of the
audience. Even if you spot the trick, you will find yourself in an awkward
position. You might for instance find yourself inhabiting what Nietzsche
calls a ‘reactive’ position, a position defined by the thing to which you are
reacting. You might for instance find yourself saying ‘there are no rules or
constraints’ or ‘there’s no such thing as syntax’ or ‘conversation has no


8

Language and social relations

structure’ or something along these lines. This would be an over-reaction.
The real issue is that if the study of language proceeds by fetishizing
restricted data about fragments of language the possibility that such a
study could reveal something about social relations among persons across
diverse languages and cultures simply vanishes. A better response is to
locate the narrowed purview within a wider one. To observe, for example,
that when syntacticians claim to describe the concatenation rules of a
‘language’ they are not describing a language at all, but only a socially
locatable register of a language (often the register called ‘the Standard
Language’), and the question of how they come to have any particular
intuitions about it is part of what a social theory of language must explain.
Or to observe that when the role of discourse in society is approached from
the standpoint of some specific genre, such as ‘face to face conversation,’
the models identified as models of discourse make opaque discursive
processes that connect persons at different scales of social grouping and
historical time through that conversational encounter, but also through
encounters whose genre characteristics are entirely different. An even
better response is to make explicit the limits within which specific theories
of language can explain aspects of it, so that the fruits of attachment to
singular ideals can be enjoyed without nearby fields falling fallow. These
are issues I take up in more detail later, especially in Chapters 1 and 2.
We shall do better to think of semiotic norms of language not as rules or
constraints but as conditions on the construal of messages as signs. Such
conditions are only satisfied for persons for whom these messages function
as signs. You may not know the language your interlocutors are using. Or
you may know it quite well, but speak a different register of it, and be
inclined to call the register they are using by a specific name (‘legalese’ or
‘baby talk,’ for instance) and get only part of their gist. Every such register

of a language has a describable grammar, which may differ only fractionally from Standard register, if a Standard exists, and only in some limited
structural realm, such as lexicon or phonology; but this fractional difference itself conveys social information, is itself diacritic of social contrasts,
which may also become commodified in various ways, even named as
emblems of distinct social identities. Issues of register difference are discussed in Chapter 3. The social life of such commodity forms is the main
focus of Chapter 4. And issues pertaining to social diacritics, emblems and
identities is the topic of Chapter 5.
Reflexive operations can fractionally transform a norm, and such operations can recursively be iterated through further semiotic activity. This
point is implicit in what I said earlier about the dialectic of norm and trope.
Much of the complexity of the ways in which language can clarify social
relations for users derives from the capacity of language users to acquire a
reflexive grasp of particular aspects of a semiotic norm – what the norm is,